The primary goals of the proposed study are to (a) develop an interdisciplinary network addressing the integration of two different methodologies for functional neuroimaging, magnetic source imaging (MSI) and functional magnetic resonance imaging (fMRI) in order to (b) evaluate a model of the cerebral mechanisms that support skilled reading. These methods differ in their sensitivity to the spatial and temporal components involved when the brain is activated while performing a cognitive task. By combining the strengths of the University of Texas-Houston MSI laboratory and the resources for fMRI at the Haskins Laboratories/Yale University, both of which have published extensively on the neural mechanisms underlying reading, we will integrate these two functional neuroimaging modalities and develop a more comprehensive model of the cerebral mechanisms underlying skilled reading. This model stipulates that printed word recognition is related to the development of a highly organized cortical system that integrates orthographic, phonological and lexical-semantic features of words. This system involves two posterior circuits in the left hemisphere (LH): a dorsal (temporo-parietal) and a ventral (occipito-temporal) circuit, along with a third circuit (inferior frontal gyms). Each circuit has a different role in reading, varying not only spatially, but also temporally. A series of experiments are proposed involving 30 healthy adults who would receive both MSI and fMRI imaging modalities while performing different reading tasks in order to (a) determine precisely which brain areas are functional components of the neural circuits supporting reading; (b) determine how the three putative neural circuits differ from one another in terms of the types of information processing performed by each. We will systematically modulate demands on different component processes in word recognition to reveal differences in the kinds of information each circuit processes; and (c) determine how the three circuits interact with each other in real time to support by integrating three types of data: spatial extent of regional activation, relative timing of the engagement of different brain areas, and functional connectivity. We also propose cross training in the two technologies and the development of an interdisciplinary network composed of investigators at both institutions. Together these efforts will expand the interpretability of imaging data by direct comparison and integration of the techniques and lead to a more elaborated theory of the cerebral mechanisms underlying reading.